Something I keep hitting with AI agents that browse: the browsing itself eats the budget. In an MCP client like Claude Desktop, every look at a page sends back the whole tree, so after two or three pages the context is full and the agent slows to a crawl. A simple task turns into a dozen round trips.
I want to hear how others handle this:
- If your agent drives a browser, how many pages before the context gets tight?
A local MCP server for Claude Desktop, Cursor, and any MCP client. Browsing tasks finish ~3.5x faster than Playwright MCP, run in parallel, and cost a fraction of the tokens. Pages stay on your machine.
